Multivariate analysis of the surface water in tropical river-estuary-sea continuums with changing land use.

Abstract

The monitoring and assessment of surface water properties has been mainly conducted for freshwater rivers using multivariate analyses. The present study aimed to evaluate the effectiveness of introducing the absorption spectrum of water in order to develop a better monitoring and assessment technique for the surface waters of a river-estuary-sea continuum. Fourteen variables were measured at 17 different sites along three rivers, including estuaries and coastal sea, in Brunei Darussalam, Borneo, from August 2017 to December 2020. Among the 17 sites, freshwater streams surrounded by primary forests and coastal seawaters had the lowest ranges of the absorption coefficient at 350nm, suspended solids, and nutrients, but had the highest range of dissolved oxygen concentration. The other sites were more or less affected by natural leaching of chromophoric dissolved organic matter (CDOM) from forests, soil runoff from agricultural lands, and wastewater discharge from urban and industrial activities. The principal component analysis (PCA) showed that four principal components were significant and explained 91% of the total variance in the dataset. Four most explanatory factors were also extracted using the factor analysis (FA). In addition to the absorption spectrum, only seven basic water variables (salinity, pH, dissolved oxygen, total suspended solid, particulate organic matter, particulate inorganic matter, phosphate) were found to be necessary to assess the spatial variations in water properties at the present river-estuary-sea continuum. We conclude that it is important to include absorption spectrum data in water monitoring programs for more efficient and comprehensive assessment at land-sea interfaces.